George Bosilca 87299e0b1c Get rid of the division in the critical path. ... Amazing how a bad instruction scheduling can have such a drastic impact on the code performance. With this change, the get a boost of at least 50% on the performance of data with a small blocklen and/or count. Signed-off-by: George Bosilca <bosilca@icl.utk.edu>		2019-08-05 09:39:44 -04:00
..
class	opal/free_list: fix race condition	2018-10-16 15:28:20 -06:00
datatype	Get rid of the division in the critical path.	2019-08-05 09:39:44 -04:00
dss	Add OPAL_VPID to unpacking	2018-11-21 11:48:58 -06:00
etc	Correct the comment in the default MCA param template - we do not support a param called "component_path". The correct syntax is "mca_base_component_path"	2018-01-05 08:46:44 -08:00
include	opal/atomics: Add acquire semantics back for spinlocks	2019-03-19 19:45:20 +05:30
mca	btl/openib: fix issue 6785	2019-07-12 08:21:21 -06:00
memoryhooks	opal: rename opal_atomic_init to opal_atomic_lock_init	2017-08-07 14:15:11 -06:00
runtime	opal/progress: protect against multiple threads in event base	2018-09-21 14:40:08 -05:00
test/reachable	reachable: add tests	2017-09-19 19:42:54 -07:00
threads	Fix the leak of fragments for persistent sends.	2019-05-03 06:20:02 -07:00
tools	Revert "Update to sync with OMPI master and cleanup to build"	2016-11-22 15:03:20 -08:00
util	snprintf() length fix for info	2018-09-21 14:47:11 -05:00
win32	opal: standardize on max hostname length	2016-04-24 08:19:47 +02:00
common_sym_whitelist.txt	opal: add code patcher framework	2016-04-13 17:16:13 -06:00
Makefile.am	opal: remove generated asm code	2017-08-03 09:18:58 -06:00
win_makefile	Purge whitespace from the repo	2015-06-23 20:59:57 -07:00